Managing Performance and Long Range in BeVs (Technical Questions)

[lolachampcar]

The more I chew on the whole P85D, L upgrade, P90DL and performance thing the more I'm under the impression that Tesla is entering yet more uncharted territory.

I'm beginning to think that Tesla is defining impulse power from their packs now that they have motor(s) that way far exceed the packs ability to deliver power. Tesla has chosen the path of producing long range electric vehicles using, in part, performance to sell them. These are some of the observations that have be curious-

I drove my P85D to the Service Center to test drive a P90DL. I made a point of leaving battery conditioning off as I did not think it applied here in sunny south Florida and the DragTimes guys and others have pretty much determined it does nothing in warmer climates. Anyway, the P90DL felt much quicker then my P85D in short sprints from 30, 40, 60 and 70 mph. I made a point of doing the exact same sprints in my fully charged P85D on the way to the test drive and on the way home to get a back to back to back feeling for the two cars. The difference I felt should exceed the couple of tenths difference the 1/4 mile data would suggest from 40-100 mph.

Tesla surely thought they could deliver on the full P85D performance described at launch by the time it came to ship the product. It would seem they failed to do so with the higher end performance and thus the tweet and web page announcement of an OTA update to provide performance at speed not seen outside the factory. That OTA update became the L hardware upgrade.

The P90DL has yet to meet the quoted 1/4 mile time in production cars. It appears possible as MotorTrend has tested a car that reliably gets 10s (their words, not mine). Whatever the secret sauce, it has yet to spread amongst current P90DL owners.

The P90DLs MotorTrend 1/4 mile times are very good but they are not in agreement with the trap speeds. It would seem power is really being pulled at the end of the run, even more so then the end of a P85D run.

All of the above points to Tesla trying to pull off an incredibly difficult balancing act. It also has me wanting to think further outside the ICE box when trying to understand BeV performance.

Some forum members are laser focused on the one gear thing being a limiting factor in high end performance. This theory does not jive well with my experience with direct drive RC cars, planes and helicopters that have excellent high speed performance while retaining the capability to strip the plastic gears they use if the soft start function is not employed. There are also numerous electric drag bikes and cars that use a single gear, launch incredibly hard yet have very high trap speeds. In short, the one gear thing may have some impact but I do not believe it is controlling.

My thinking is starting to migrate towards impulse capability. If my butt dyno is anywhere near right from my P90DL test drive, the 90DL is capable of doing better 40-80 mph when rolling on from a steady state 40 then it is passing through that range in a 1/4 mile run. This is slightly similar to an air to water intercooler system where the ICE can produce large amounts of power until the cold water in the system is warmed. Once the initial slug of energy is pushed into the water, the charge cooling system efficiency falls off dramatically as does engine power. The battery system may be doing something similar but at a MUCH faster rate. For longevity's sake, you may only be able to pull an XYZ slug of energy out of the pack at any one time without doing damage.

If the above is the case, it becomes even harder to truly understand and quantify MS' performance. If anyone has a VBox in the south Florida area, it might be worth doing a bit of testing to see if acceleration numbers are indeed different depending upon the amount of energy that is being extracted during a single run.

 

[lolachampcar]

Now that we have 700ish hp worth of motor available to us, I would really like to see some high discharge rate capacity added to a P85D such that 470 KW of draw could be maintained for a fixed amount of time and for a certain number of times. I've not digested the REST data but lets say the battery can provide 350 KW continuous for the bursts we are talking about. That leaves a 120 KW gap to fill in if we want to use all the available motor power. If we want ten usages of ten seconds each we are looking at 100 seconds or just under 3.5 KW-Hr of capacity. That size battery would require a 35C discharge rate to fill in the 120 KW hole which is within the capabilities of modern RC batteries.

To accomplish the above, you would need to manage pack voltages such that the filler pack started supplying current when the primary pack droops. A contactor would be required to isolate the filler pack when not employed as pack voltage differences would drive cross charging/discharging when the motors were not sucking up all the current.

The engineering time would be expensive but the hardware costs would be minor. The filler pack could charge slowly off the primary or during regen when you are not beating on the car. Now is Saleen were truly capable of real BeV hop up engineering..............

 

[techmaven]

I think Tesla is more focussed on long term reliability rather than high performance. It's just that the result of their electric drivetrain gives them high performance relative to ICE competitors. I bet there is a maximum impulse power that Tesla is willing to subject the 85 kWh and 90 kWh battery too without significantly compromising long term life. If you get better at using that power earlier, you have less of it for later. The pack fuse precision necessitating the ludicrous upgrade hardware was just one of the issues, probably the critical path issue when looking at the P85D's performance. But then we run into the total impulse power issue which can be solved in several ways... increase the size of the pack to reduce the c-rate of any individual cell, change the battery chemistry to be more tolerant of high c-rate discharge (which usually have a negative effect on specific energy of the same technology) or just allow more damage to occur.

In the ICE sports car world, allowing more damage to occur is surprisingly commonplace. Witness the GTR's launch control and the resulting effect on the transmission's lifespan. I wonder if Tesla could just supply a software "upgrade" that allows that damage to occur, but voids or significantly reduces the drivetrain warranty and sell it as a track pack option for 4-5 figures? I think that would be on par with some ICE manufacturers and the market. After all, 5 figure overhauls of sports car drivetrains in years 6-10 seems to be relatively commonplace.

If Tesla was really focussed on high performance, they could offer a 60 kWh pack that can make Supercharger jumps but tolerates a far higher discharge c-rate for roughly same price as the 90 kWh pack. The resulting mass is likely significantly lower than the 85 or 90 kWh pack which enhances all aspects of sports car performance. They could also throw in an ultra capacitor or higher c-rate battery to take the brunt of the demands of high performance operation, taking up the rest of the frunk. Basically, their own variant of the KERS system used in racing. It might have actually made more sense, if raw performance was the goal, to go the KERS route with an ultra capacitor for $10k than the iconel fuse. I do think that the iconel fuse was designed for longevity/reliability first and not high performance, it just happens to also take out the critical path for achieving higher performance.

 

[lolachampcar]

TM,

How about the best of both worlds? Battery swap and a 10C capable 60KW battery?

 

[techmaven]

TM,

How about the best of both worlds? Battery swap and a 10C capable 60KW battery?

Sure, but then Tesla would have to build out lots of battery swap stations, a drain that competes against other priorities for their significant capex drain.

BTW, here is an interesting thread on pulse discharging and effects of high c-rate discharge on the NCR18650BE (very similar to the cell in 60/85 kWh packs) in comparison to other similar cells:

Continuous vs intermittent discharge

This little piece comparing samples of the NCR18650BE and the NCR18650BD is also interesting:
Comparing Panasonic 18650 BD vs. Panasonic 18650 BE 18650 Battery | BATTERY BRO

The BD can do 10A continuous discharge with basically the same capacity as the BE, but the charge rates are also vastly different. Would you be okay with 2/3's the supercharging rate but triple the discharge rate?

 

[3mp_kwh]

My thinking is starting to migrate towards impulse capability. If my butt dyno is anywhere near right from my P90DL test drive, the 90DL is capable of doing better 40-80 mph when rolling on from a steady state 40 then it is passing through that range in a 1/4 mile run.

Remains untested, and something that could change the will to go through with Ludicrous upgrades. My own experience with battery voltages (cars, bike lights, RC cars) is that voltage sags, with load (simple, I know). Unloaded, whether the pack is carved to fewer batteries doing the job, or all of them, the starting voltage @40mph is likely higher than if passing by 40mph during a quarter mile run. It could also be the timing of a high C-rate, and taper that the car requires as function of duration + amperage (allowable heat).

Apart from the electrical talk, it would be good to know if the 40-80 is faster when not from a launch.

If Tesla was really focussed on high performance, they...

would hire a professional driver, and develop a drivers mode at little to no hardware cost.

 

[techmaven]

would hire a professional driver, and develop a drivers mode at little to no hardware cost.

Does Chris Porritt, ex-Aston Martin Chief Engineer that designed the Aston Martin One-77 count? He was hired by Tesla in 2013.

 

[AB4EJ]

Now that we have 700ish hp worth of motor available to us, I would really like to see some high discharge rate capacity added to a P85D such that 470 KW of draw could be maintained for a fixed amount of time and for a certain number of times. I've not digested the REST data but lets say the battery can provide 350 KW continuous for the bursts we are talking about. That leaves a 120 KW gap to fill in if we want to use all the available motor power. If we want ten usages of ten seconds each we are looking at 100 seconds or just under 3.5 KW-Hr of capacity. That size battery would require a 35C discharge rate to fill in the 120 KW hole which is within the capabilities of modern RC batteries.

To accomplish the above, you would need to manage pack voltages such that the filler pack started supplying current when the primary pack droops. A contactor would be required to isolate the filler pack when not employed as pack voltage differences would drive cross charging/discharging when the motors were not sucking up all the current.

The engineering time would be expensive but the hardware costs would be minor. The filler pack could charge slowly off the primary or during regen when you are not beating on the car. Now is Saleen were truly capable of real BeV hop up engineering..............

This doesn't necessarily require an auxiliary battery; due to the short times the extra power is required, we might consider using a capacitor-based system for the boost. It could be charged when power is available, and a capacitor can provide current output (for short bursts) at whatever current you want, up to the capacity limit of the inverter.

 

[3mp_kwh]

Does Chris Porritt, ex-Aston Martin Chief Engineer that designed the Aston Martin One-77 count? He was hired by Tesla in 2013.

Then maybe all they need are his notes! Putting all the driver control in the hands of many drivers who can manage it is something I think is at cross-purposes with designing a safety oriented traction control system. I'd emphasize the limiters come really late, where I sincerely doubt the majority of PD drivers go, but for those of us who sense there is so much "left in the can" it would be great, I think. Use the car's limits, when you want to control it. Turn the nannies on, when you don't (in rain, late night, etc.). You need more than one mode, in this tech age.